It seems like some data researchers from Facebook are now auditioning for the role of Yenta from “Fiddler on the Roof” for over 1 billion people that subscribe to this very popular social networking site. Now, before you start singing “Matchmaker,” you might want to know how these data scientists are determining your love interests for you.

One of the first methods that data researchers use to determine that special someone is called “embeddedness.” In terms of this study, embeddedness refers to looking at how many mutual friends you have with another person. This method is used to determine the degree of closeness between two people based on the number of their mutual friends. 

Well, I can tell you from personal experience that the person I have the most mutual friends with is not my lover. And although 24.7 percent of the time embeddedness predicted an individual’s significant other, there’s a larger portion of people that this method did not work for.

To start, there is a major issue with this method that I am surprised the researchers did not discuss. What happens when the person who you share the most mutual friends with is your best friend? 

For some, it may develop into a long and loving marriage, but others may realize that they should have just stayed friends. That could end up as a pretty rocky friendship. If popular media and personal observations serve me right, then dating your best friend is practically a crapshoot.  

However, there was one method that was far more successful than embededdness called “dispersion.” This method relies on you and your predicted “lover” having a high amount of mutual friends in the same networks. In other words, this method looks at mutual friends who went to the same high school, university or work at the same company as you.

This method makes a little more sense. Instead of just using mutual friends from your high school that you may have attended five or six years ago to determine your match, now they are comparing the number of mutual friends that continue through college and your career. These data researchers seem to be piecing together your romantic timeline straight from your Facebook timeline, so to speak. 

What differentiates dispersion from embeddedness is embeddedness just looks collectively at your mutual friends, but dispersion looks at the amount of mutual friends that you have in specific networks.

The success of the dispersion method is quite astounding. While the embeddedness method only garnered 24.7 percent accuracy, the dispersion method was accurate 50 percent of the time. Additionally, when researchers used the dispersion strategy on married couples it predicted what couples were married 60 percent of the time.

What makes this process even more interesting is its ability to get around the coveted “Facebook official” label. This pre-teen validation may have no significance as these researchers continue to refine their methods because they are looking at the number of mutual friends in your networks rather than Facebook posts — such as comments, statuses, pictures and likes. 

So, that picture of you making out with Jessica is not the only indicator that you two are clearly in it for the long run. Now instead of having researchers look at your Facebook friends list, just do it yourself.

Dispersion can even determine how healthy a relationship is. The more dispersed a relationship is, or the more integrated each person is within each others’ networks, the stronger the two are romantically connected. 

The degree of dispersion determines what relationships will still be maintained in 60 days. Although this finding was only in regards to non-married couples, seeing the results based on the degree of dispersion for married couples could possibly determine the strength of current marriages. 

All of this makes me wonder, is this our future? Data researchers are ultimately pulling together numbers on spreadsheets and telling us who we are most compatible with according to their calculations. This theory already exists in another form, online dating.

Sites like eHarmony and Match.com are large vestibules for those who are taking a chance on the single men and women who put themselves out there on the Internet. However, these sites use personality-based algorithms and research methods to determine who is compatible with who, and they allow the user to select who they like the most.

With embeddedness and dispersion, there is no personality comparison, only a method comparing numbers of friends that, for the most part, are barely that. I have friends who have almost 2,000 Facebook friends, but I highly doubt that they can say that each one of those people is truly their friend. 

While this method seems to be pretty accurate, it is still far from becoming the proverbial Yenta of the Internet. As these algorithms become more advanced and accurate, there will always be those couples that are unpredictable. No one will ever truly know how they got together and how they maintained a lasting, loving relationship.

Max is a freshman in DGS. He can be reached at [email protected].